England and Wales High Court (Technology and Construction Court) Decisions

You are here: BAILII >> Databases >> England and Wales High Court (Technology and Construction Court) Decisions >> Cadcam Technology Ltd v. Proel SRL, Midwest Europe Ltd [2000] EWHC Technology 63 (29th September, 2000)
URL: http://www.bailii.org/ew/cases/EWHC/TCC/2000/63.html
Cite as: [2000] EWHC Technology 63

[New search] [Help]

Cadcam Technology Ltd v. Proel SRL, Midwest Europe Ltd [2000] EWHC Technology 63 (29th September, 2000)

IN THE PATENTS COUNTY COURT

Claim No. 99 PAT 007

Mr Peter Prescott QC

Cadcam Technology Limited

Claimant

and

(1) Proel SRL

(2) Midwest Europe Limited

Defendants

Hearing date 13, 14 and 15 November 2000 at

The Central London Patents County Court

JUDGMENT

I direct pursuant to CPR Part 39 P.D. that no official shorthand note shall be taken of this judgment and that copies of this version as handed down may be treated as authentic.

Peter Prescott QC

29 November 2000

Brief history of the case.

1. In 1996 European Patent 0,641,171 B1 was granted to the Claimant. Its title is "Cutting and Embroidery Process". It claims priority from 16 May 1992. On 14 July 1999 the Claimant sued the Defendants in this court for infringing the patent. The name of the First Defendant was given as "Proel SpA", but by common consent it should be "Proel SRL", and I shall direct accordingly. The complaint was that the Defendants were supplying software known as LASERCUT and computer-controlled laser cutting machines upon which such software operates.

2. For present purposes the only claim of the patent that matters is claim 1. No other claim is alleged to be infringed yet to have independent inventive significance.

3. The Defendants deny that what they are doing falls within the scope of the patent even when considered together with the law of contributory infringement. They also allege that that the invention of claim 1 was anticipated by a prior user. They further allege that claim 1 is anyway invalid for obviousness having regard to common general knowledge or information communicated to the public by acts which are well established.

4. For reasons I have given in the course of a preliminary ruling, I have decided to take one aspect of the case as a preliminary issue. If decided in favour of the Defendants it will dispose of the claim against them. The aspect I refer to is dual, namely, the correct construction of claim 1 of the patent, and the validity thereof having regard to prior art which is well established or is common general knowledge. (The pleaded prior user does not fall into that category for its nature is disputed.) The case was originally estimated to last 5 days. By the time of the trial this had risen to 9 days. In the event, the trial of the preliminary issues has lasted 3 days.

The witnesses.

5. A number of witness statements and expert reports were filed by the parties. Some were entirely or partly devoted to issues of prior user and/or contributory infringement, issues which I am not deciding at this stage. Others were eventually agreed to be unnecessarily duplicative, or irrelevant to the issues I do have to decide. Those that remain for present purposes are:

For the Claimant: the 1st witness statement of John Bell; the witness statement of Steven F. Widincamp; the 1st expert witness statement of Dr David White.

For the Defendants: the expert report of Claudio Gremese.

I have read the other witness statements but I am satisfied that they are not really material to what I now have to decide.

6. With the somewhat doubtful exception of Mr Widincamp, none of the above-named witnesses was truly independent in terms of business interest or employment. In itself that is no objection provided the witness is fair and impartial. In my judgement Mr Gremese achieved that standard on the whole, but Mr Widincamp, Mr Bell and Dr White did not. Mr Widincamp was too much of a salesman and I felt that he was inclined to be evasive when it suited him. Mr Bell was impressive from one point of view, in that he had the interests of his company very much at heart and promoted them extremely well, but he was argumentative, not impartial, and was prepared to prevaricate on at least one occasion. Those persons were called as witnesses of fact. Dr White was called as an expert. In my judgement he did not comply with the rules which apply to the preparation of expert reports. Those rules are intended to ensure that an expert witness regards his first duty as being to the court, not his client or his employer. Because Dr White did not approach the writing of his report in that spirit I do not regard it as reliable. However, once Dr White came to appreciate his proper role he much improved. If only he had approached his task like that from the beginning. Once he had reformed he gave evidence which I think is reliable, except on one occasion when he reverted to what I can only describe as the 'party line', perhaps understandably so having regard to the difficult situation he was in.

7. All these matters were brought out in the admirable cross-examinations conducted by Mr Richard Davis for the Defendants, who was fair, measured and to the point.

The Defendants' witness Mr Gremese was produced as an expert. Like the Claimant's witnesses, he was not independent, indeed he had been a partner in Proel for 20 years. Despite this I prefer his testimony whenever it conflicts with that of the others. There were some language difficulties, but they were not great, and he gave his answers in satisfactory English. In my judgement his expert report was reliable, subject to some errors which he corrected in cross-examination.

8. Apart from the above sources of evidence, both parties relied on certain hearsay evidence concerning the availability of machinery at the priority date; the source being the findings of fact of Jacob J in Gerber Garment Technology Inc v. Lectra Systems Ltd [1995] RPC 383 at pages 389 (Defendants) and 390 line 30 to 391 line 9 (Claimant).

9. I have to construe claim 1 of the patent and decide whether the invention claimed therein would have been obvious to a person skilled in the art on 16 May 1992 in the light of common general knowledge. The patent has to be read purposively, as if through the eyes of a person skilled in the art. I shall therefore review the industrial background.

The industrial background.

10. Embroidery is the art of decorating fabric with needlework. The work can be laborious and during the 1860s machines were introduced that could stitch the required pattern automatically. They were programmed, and the program consisted of a punched card. The holes in the card told the machine how to move to produce the required pattern of needle-stitches. It seems that the originator was a Swiss called Isaak Groebli. He invented one of the two main types of machine which are still used, the so-called Schiffli machine. This is a large machine with a vertical frame. Groebli himself was not the first to think of using punched cards for programming machines. A machine for weaving cloth, programmed by a punched card, had already been perfected by the end of the 18th century by Jacquard, whose name is now a dictionary word.

11. The other main type of embroidery machine is known as the Multi-head, introduced in the early part of the 20th century as being better suited to short orders. It is generally smaller than the Schiffli, works horizontally and consists of a multiplicity of sewing-type machines. Like the Schiffli, it was programmed by the use of a punched card or tape.

12. Thus the weaving and embroidery trades used the idea of programming a machine to perform a repetitive task long before the computer manufacturers. Nowadays embroidery machines are often controlled by electronic computers and for this purpose the requisite pattern is digitised. The digitised data may be acquired by scanning a design drawing; or the design may be created directly with the assistance of a computer (CAD). However it is done, the process of inputting the data is often called punching, even to this day.

13. Appliqué is ornamental work in which fabric is cut out and attached, usually by sewing, to the surface of another fabric to form pictures or patterns. The shape of the cut-out piece may provide the desired visual image just by itself. But very often the image is enhanced by decorative needlework (embroidery). Exhibit D1 is an illustration. A piece of brown cloth has been cut out in the outline of a bear. It has then been attached to the surface of another fabric by sewing, and further needlework has been added to simulate its paws, legs, eyes, ears and so forth.

Although I have referred to the use of 'fabric' (which strictly speaking implies a woven cloth) the material could be vinyl or leather.

14. So far as I have been told, appliqué pieces are still positioned on the base fabric by hand. Thus the operative must be supplied with a location guide. This is done by causing the embroidery machine to lay down a mark-out pattern. It consists of the outline of the cut-out appliqué piece. Often, but not always, the outline will represent of the entire contour of the piece. In order to hold down the appliqué piece temporarily adhesive is used.

15. Because the process of applying the decorative stitching to the appliqué may cause it to distort, the cut-out appliqué piece should first be secured firmly to the base fabric. This is done by applying a hold-down (or fixing) stitching pattern – it may zigzag along the edge of the appliqué piece.

Thus in appliqué work there may be three kinds of stitches: mark-out stitches, hold-down stitches, and purely decorative stitches.

16. By 1992 there were two standard methods which were used for producing appliqué pieces of the right size and shape. One method was to make them over-size and to trim them by hand once the basic stitching had been laid down. This was labour-intensive. The other method was to pre-cut the pieces to size by the use of a stamping die. The die consisted of a knife blade bent to the required shape and supported in a wooden board. This was used to chop out the appliqué element from whole cloth. The manufacture of these dies was a separate trade. Because each shape of appliqué piece would require a die to be made to order the method was suited to longer production runs. (If it were a question of short runs the hand-trimming process would be preferred.) In order to make the die the manufacturer needed to be supplied with the requisite information, that is to say, the shape of the desired outline. Commonly, he would be supplied with a design drawing.

17. Because there was a practical limit to how accurately the knife blade could be bent, it was not always possible to make dies which would cut the cloth precisely. This could be so where the shape was intricate or had unusually sharp bends. Also, it was not unknown for the die to be imperfect because the design drawing had been interpreted incorrectly, or inconsistently in relation to the stitching pattern. Furthermore, if the embroidery was to be done on a Schiffli machine it was not possible to use die-cutting to get appliqué pieces of the precise shape, although this was not so in the case of Multi-head machines. In practice, the parties to this case are concerned with appliqué work embroidered by the use of Multi-head machines.

18. There are not many technical textbooks on embroidery but one such is The Art of Embroidery in the 90's by Coleman Schneider, a person of repute in the industry. An edition was published in 1992. A copy is in evidence (exhibit D2). The copy was presented by the author to Claudio Gremese, a witness in this case, on 27 April 1992, shortly before the priority date of the patent, and its flyleaf is inscribed accordingly..

The following passage occurs at pages 216-7:

APPLIQUES: For Schiffli, appliques are applied after the direct or basic stitching is completed. They are usually in the form of cut pieces large enough to cover the applique area completely. Die cut pieces of the exact size are impossible for Schiffli. The pieces are pinned onto the spanned goods. In some cases the applique for a 4" figure might be 6" wide, the length of the machine. After the goods is off-spanned it is hand cut as close as possible, but unless laser cut it will never be entirely clean, and if woven goods are used the possibility of fraying always exists.

For Multi-head machines the pieces can be die cut. The pattern should contain a stitched outline over which the applique will be fitted. A stickum of some type, heat seal or pressure sensitive will hold it in place. The pattern is programmed to stitch an outline around the shape to hold it in place while the edge is stitched. Unlike the Schiffli above, the embroidered steil will be stitched over the edge, making it perfectly clean without any fabric showing along it.

19. The same book contains a supplement written because, so said its author, "The advancement of Embroidery technology has been so rapid that this book had to be updated 2 weeks after it was off the press". At the end there is a section on computer-controlled laser cutters. The last sentence (page 462) says:

It is the wave of the future, and with refinements this is the way emblems, appliques and labels will be cut in the future.

However such machines were very expensive in May 1992. There were other machines which could cut cloth, e.g. by the use of knife blades or water-jets. For various reasons they were not really satisfactory for present purposes.

The patent.

20. The title of the patent is "Cutting and Embroidery Process". It is not a complex document and the gist of it could be understood by any person with a normal secondary education. Even so, it must be read through the eyes of a person skilled in the art for he will know things which the ordinary reader may not. Further, claim 1 is not at all easy to interpret.

21. The patent begins as follows.

'This invention relates to the control of cutting and embroidery machines and particularly to the control of such machines for reproducing designs incorporating appliqué elements'.

Background of the invention.

Computer control of embroidery machines is well known. Control may be direct or via a program stored on a punched tape ... Such computer controlled embroidery machines have been used for generating designs incorporating appliqué elements, for example in which shaped elements of one sheet material (the appliqué elements) are attached to a base of the same or a different sheet material by stitching in the embroidery machine.

The sequence of steps in generating prior art appliqué designs has been as follows. First a representation of the desired design is entered into the computer by means of a design program or a video input or by other well-known methods. The computer, under the direction of an operator, uses the representation of the design to derive stitch command data for the embroidery machine. The operator may in certain cases manipulate the design, for example by rotation or enlargement, before the stitch command data are generated. A sufficiently large sheet of the appliqué material is placed over the base material in the embroidery machine and, in response to the stitch command data from the computer, the embroidery machine stitches through both layers an outline of the desired shape of the appliqué element. A human operator then trims the appliqué material around this outline and, once trimming is complete, further stitching may be carried out under computer control, for example to hide the edges of the appliqué element or to add further detail.

The labour-intensive nature of the manual trimming operation described above is clear. The more complex the outline of an appliqué element, the more difficult is the trimming operation and where the outline is not a simple closed curve, for example with a ring-shaped appliqué element, the above procedure may not be practical at all. Furthermore, it is very wasteful of material to begin each appliqué operation with a plain sheet of appliqué material, much of which must subsequently be thrown away.

22. It will be noticed that the patent does not give a wholly accurate account of existing practice. A person reading it with no previous knowledge of the industry would be led to suppose that there was no alternative to the laborious step of hand-trimming. In my judgement those skilled in the art would have known that in appropriate circumstances appliqué pieces were pre-cut to size by the use of dies.

23. There follows a summary of the invention in the language of claim 1. I shall consider claim 1 later in this judgement. The patent proceeds to mention a preferred process which includes the further step of stitching a mark-out pattern on the base material in accordance with the stitch command data, before the steps of transferring the cut-out appliqué element to the base material and stitching the appliqué element to the base material. As I have observed, the mark-out pattern is for the benefit of the manual operative and indicates to him the correct location of the appliqué element.

24. After referring to certain other matters, the patent continues:-

The cutting machine for cutting the appliqué element from the sheet material may be a laser cutter which cuts through the appliqué element from above. Alternatively a suitable bladed cutter can be used instead of a laser cutter. Laser cutters are very high precision tools and may be set, if desired, to cut through the both the appliqué element and its backing element or alternatively, if desired, to cut through the appliqué element but not the backing layer.

25. As the patent has already explained, the 'backing layer' in question is a sheet to which the appliqué element is temporarily stuck but from which it can be peeled when required – rather like the sheets of labels which we use for floppy disks. Thus the patent presupposes that there are available computer-controlled laser cutters of such exquisite precision that they can cut through the appliqué fabric yet not through its backing. It also presupposes that cutting machines are available of such precision that

the appliqué element is cut to the desired shape before being transferred to the base material ... The need to trim around the element is completely eliminated ...

26. There are no instructions how to make a computer-controlled cutting machine which can do these things and so the patent presupposes that the skilled reader will know how to go out and get one in commerce. In my judgement that would have been correct at the priority date of the patent. However at that time such machines were very expensive, as I have said. A die was very much cheaper to procure although of course a separate die had to be made for each design. There is no doubt that satisfactory computer-controlled embroidery machines were available.

27. At column 3 lines 16-38 the patent states as follows (the emphases are mine):

Generally the analysis of the design to derive the data from which the stitch command data and cutting command data are generated will be carried out in a computer. Because the computer can control ... both the cutting machine and the embroidery machine using similar software, the process is very efficient. The data representing the design can be easily manipulated in the computer to alter the design before the stitch ... and cutting command data are generated. The appliqué element is cut to the required shape before being transferred to the base material, and the embroidery machine can be controlled to mark out the same shape or relevant parts of the shape on the base material prior to transfer of the appliqué element, which makes location of the appliqué element extremely quick and simple, particularly when it is backed by adhesive. The need to trim around the element is completely eliminated so the process is much faster and more complicatedly shaped elements can be used than with the prior art process. The cutting machine can be controlled so that the layout of the elements cut from the sheet material gives rise to little wastage.

28. Claim 1 of the patent contains no explicit statement about the use of computers and the Defendants referred to the above passage in support of their argument that the claim does not require the analysis of the design (which is required in order to derive the stitch command data and the cutting command data) to be performed on a computer at all, pointing to the word 'generally'. Thus the use of a computer is a preferred option but is not essential. In my judgement it would not be very satisfactory to place much weight on this point. The passage is open to another interpretation. Namely, that although, in general, the derivation of the data by analysis of the design is done on the computer, certain aspects may involve human manipulation.

29. There then follows a somewhat more detailed description of the invention with reference to the drawings. In summary, the design to be reproduced is entered into a computer, either directly by a design program (i.e. CAD) or by acquiring the design by means of a video camera or other well-known means. Using the computer and certain unspecified software, command data is derived for controlling a cutting machine and an embroidery machine. The stitch command data includes the mark-out stitching. It also includes hold-down (fixing) stitching. Those preparations having been completed, the computer sends the cutting command data to the cutting machine and while the appliqué element is cut out the computer sends the stitch command data to the embroidery machine to lay down the mark-out pattern. A human operative locates the cut-out element to the mark-out pattern by hand, where it is held in place by adhesive. The fixing stitches are then applied by the machine in obedience to the stitch command data.

30. The Claimant's case is that it was inventive to have an integrated system which would analyse the design drawing to derive the command data for the cutting and stitching machines respectively. It was said that a benefit of doing this was that the outline of the cut-out appliqué element, and the stitching to be applied along that outline, would coincide in a reliable manner. It was also said that analysing the design drawing to get the right commands was no easy matter, indeed was open to a degree of 'artistic interpretation'.

31. Although I have summarised the above description in short terms, the description itself contains little extra detail which is pertinent for present purposes. There is no concrete description of any software and next to no information about its logic is provided. It is not clear to what extent the software is supposed to be able to derive the requisite data automatically, that is to say, without some form of intelligent human input. It would be fair to quote the relevant passages in full:

Once entered into the computer 2, the design 12 may be manipulated using conventional processing techniques under the control of a human operator, for example to change its size, proportions or orientation. The basic design is processed in various ways using software to generate command data for controlling the cutting and embroidery machines.

The first processing step is to select the outline ... of the design 12 that will be cut out in the appliqué material. The software for this purpose includes means for line and curve entry and for text generation (so that the appliqué elements may easily be made characters of text).

Once the operator is content with the form of the cutting outline for the appliqué element, certain components of the design outline are selected for use as mark-out guides. From these selected components, the computer 2 generates stitch command data to be sent to the embroidery machine 10. These data control the machine 10 to produce a pattern of stitches 14 on the base material 16, which stitches indicate the correct location for the appliqué element 18 ...

The next stage of the process is for the operator to use the design processing software of the computer 2 to determine the desired layout of the fixing stitches 20 that will hold the appliqué element 18 to the base material 16 ... The desired layout and sequence of fixing stitches 20 having been selected, the computer generates from data representing the design 12 a suitable further sequence of stitch command data, schematically illustrated at 22 in Figure 1.

32. What is schematically illustrated in Figure 1 is simply a stitch pattern zigzagging along the edge of the appliqué element, which element has the shape of a stylised letter 'A'.

I believe the above passages are consistent with a process according to which the various aspects of the design (including the cutting outline of the appliqué element and the positioning of the mark-out and fixing stitches) are created by the intellect of a human operator, the computer merely behaving as a tool for drawing, as in CAD, and wherein the only 'analysis' done by the computer is to measure the appropriate values e.g. co-ordinates. Be that as it may, no further information of any relevance is provided in the patent, and if the computer is to do more the creation of the necessary logic is left as an exercise to the skilled reader.

33. I now turn to claim 1, but it will also be convenient to quote claims 2 and 3 since according to the Claimant these may have a bearing on the proper interpretation of claim 1. Omitting the optional numeral references given therein (for these are for ease of understanding only and have no limiting significance in law) they read as follows:

A process for reproducing a design, comprising the steps of analysing the design to derive data as stitch command data by an embroidery machine, and as cutting command data by a cutting machine, cutting out an appliqué element from a sheet material in accordance with the cutting command data, transferring the cut-out appliqué element to a base material and stitching the appliqué element to the base material in accordance with the stitch command data.

A process according to claim 1, including the further step of stitching a mark out pattern on the base material in accordance with the stitch command data, before the steps of transferring the cut-out appliqué element to the base material and stitching the appliqué element to the base material.

A process according to claim 1 or claim 2, including the further step of applying additional decorative stitching to the appliqué element, after it has been stitched to the base material, also in accordance with the stitch command data.

34. Claim 1 has given much trouble during the case. First, because it is not easy to work out how much of it pertains to purely automatic methods and how much to processes which may be performed by the human eye, hand and intellect. Secondly, because it is not immediately apparent what are the types of stitching commands being referred to; I shall deal with that aspect first.

35. As I have said, there may be three sorts of appliqué stitches: (a) for showing the operative where to locate the appliqué element on the base sheet ('mark-out' stitches); (b) for fastening the appliqué element to the base sheet ('fixing' or 'hold-down' stitches); (c) for decorative purposes. It is apparent that claim 2 pertains to the mark-out stitches and claim 3 to decorative stitches. It is also apparent that in claim 1 the stitching of the appliqué element to the base material as specifically referred to therein is for the purpose of holding it down. However in my judgement it would be feasible in some cases to use one and the same set of stitches both for that purpose and for decorative purposes. For instance, a black appliqué cut-out element in the shape of a letter of the alphabet might be stitched down in gold decorative thread. Decoration lies in the eye of the beholder. It is also apparent that in all three claims the stitches are applied 'in accordance with the stitch command data'. It follows that the expression 'the stitch command data' is a generalised expression. It is a product of 'analysing the design' in claim 1 but the data may contain several sub-sets, namely, it must contain data for holding-down purposes and it may contain data for mark-out purposes, and data for purely decorative purposes.

36. More difficult is the question whether the various steps of claim 1 are required to be performed by automatic means. On one view the whole process could be accomplished under human control alone, without the intervention of any computer or other automatic data-processing machine. It is clear from the description of the invention (column 5 lines 18-23) that at least one step may be manual, namely, the placing of the cut-out appliqué element on the base material sheet. If one step may be manual, could others be? Indeed one branch of Mr Davis' case is that the analysis of the design to produce the requisite stitch command data and cutting command data could also be done by hand – that nothing in the claim stipulates otherwise.

37. I believe that the 'embroidery machine' is one which can automatically produce stitches to order in response to command data – this much seems apparent from the context. Likewise the 'cutting machine' is one which can cut out an appliqué element automatically in response to cutting command data. What is a more difficult question is whether the data have to be generated entirely by automatic means alone (by some computerised process, for example); or may be generated semi-automatically (by a combination of some sort of automated data-processing means and the use of the human eye, hand and intellect); or may be generated entirely by human means. Recall that programmed embroidery machines existed for more than 100 years in which the command data was present in the form of a punched card, yet in those days the card had to be prepared by an analysis of the design drawing performed by the use of human intellect and measurement alone. Of course, the holes in the card represent 'data'.

38. In my judgement claim 1 when read in the context of the patent as a whole probably requires that the 'analysing' which leads to the generation of the stitch command data and cutting command data is a reference to analysis performed with the assistance of a computer. The reason I incline to this view is because of the passages headed 'Background of the Invention' which I have quoted above. The notional skilled reader of the patent would appreciate that the command data could be obtained by human measurement and manipulation alone, as in the old days of manually-produced punched cards; but he would also realise that this would be very time-consuming, hence pointless under modern conditions. So he would probably treat the expression 'analysing the design to derive data' in claim 1 as a reference to obtaining the data with the assistance of a computer. Having said that, I do not believe it makes any difference to the result of the case.

39. I therefore interpret claim 1 as requiring a process for reproducing a design which has the following features:-

Analysing the design with the assistance of a computer to derive data for use as stitch command data by an embroidery machine.

Analysing the design with the assistance of a computer to derive data for use as cutting command data by a cutting machine.

Using the cutting command data to instruct a cutting machine to cut out an appliqué element from sheet material.

Transferring the cut-out appliqué element to a base material.

Using at least a sub-set of the stitch command data obtained in step (a) to instruct an embroidery machine to stitch the appliqué element to the base material.

40. However, Mr Tritton who appeared for the Claimant, contended that there is a further feature required by claim 1. I quote from his closing skeleton argument.

From a design loaded into a computer (see Col 4_1-3) software processes the design (Col 4_5-6) to generate stitch command data to control an embroidery machine to affix the appliqué material to the base material (which [as] is well known in the common general knowledge is around the outline) and cutting command data to control a cutting machine to cut out the appliqué (which is of course also the outline). As can be seen from Claim 2 and Claim 3, "stitch command data" can contain more than one category of stitches. In Claim 1, it will be purely the holding stitch which is stitched around the outline. In Claim 2, it will include the holding stitch and the mark out stitch. In Claim 3, it will mean those stitches and the embroidery stitches.

In effect, the concept embodied in Claim 1 is that there is an integrated uninterrupted "software data" link between the design and the cutting and embroidery machines. There may be further processing of the design (e.g. the conversion of a straight outline to a zig zag holding stitch ...) but there must be no manual intervening step at the software level so that the data must be inputted again (thus the printout of a picture will be a manual intervening step but the carrying of the data on a floppy disk is not).

41. The reference to carrying the data on a floppy disk arises from a series of questions I put to Mr Tritton in argument. I asked him whether claim 1 would cover the following imaginary scenario. Suppose the design drawing is sent to the cutting department of a factory where it is scanned and analysed to generate the cutting command data. Then the drawing is sent to the embroidery department where it is scanned and analysed to generate the embroidery command data. The respective data are used to command the cutting and embroidery machines, the other steps being in accordance with claim 1. One purpose of my question was to consider whether such a concept would have been obvious in May 1992. Mr Tritton conceded that it would have been obvious – rightly, in my judgement – but he contended that it was not covered by the claim.

42. In order to understand what he meant by an uninterrupted data link I also asked him whether claim 1 would cover the following alternative scenario. The drawing is scanned into a computer in order to digitise the data contained in the drawing. This data is then copied to two identical floppy disks which are carried by messenger to the cutting and embroidery departments, respectively. In those departments any further analysis and further processing steps as required by claim 1 are performed. Mr Tritton, if I understood him correctly, said that this alternative would be covered by claim 1 but that the concept would not have been obvious.

43. While I must read the patent as a whole, I know of no legitimate principle of construction which enables me to infer Mr Tritton's construction from the language of claim 1. In any dispute like the present the proprietor of the patent wants to give his claim a narrow meaning in the hope of avoiding a finding of anticipation or obviousness, and is therefore tempted to look for passages in the body of the specification which lend colour to his interpretation. The passages to which Mr Tritton has referred in his skeleton argument, namely in column 4, are part of a detailed description of the invention. It is not right to select such passages to suit: see e.g. Glaverbel SA v. British Coal Corp [1995] RPC 255, 269. Mr Tritton's clients would be the first to object, and rightly so, if an infringer of their patent adopted a similar technique in order to put a meaning on claim 1 which did not cover his process. In my judgement if the Claimant wished its claim to be restricted along the lines indicated above in boldface type it should have worded it accordingly; but it has not.

44. Section 125(1) of the Patents Act 1977 provides:

For the purposes of this Act an invention ... for which a patent has been granted shall ... be taken to be that specified in a claim of the specification ... as interpreted by the description and any drawings contained in that specification, and the extent of the protection conferred by a patent ... shall be determined accordingly.

Section 125 was enacted in order to conform to Article 69 of the European Patent Convention, the relevant portion of which says:

The extent of the protection conferred by a European patent ... shall be determined by the terms of the claims. Nevertheless, the description and drawings shall be used to interpret the claims.

This must be read together with the Protocol on Interpretation, which says:

Article 69 should not be interpreted in the sense that the extent of the protection conferred by the European patent is to be understood as that defined by the strict, literal meaning of the wording used in the claims, the description and drawings being employed only for the purpose of resolving an ambiguity found in the claims. Neither should it be interpreted in the sense that the claims serve only as a guideline and that the actual protection conferred may extend to what, from a consideration of the description and drawings by a person skilled in the art, the patentee has contemplated. On the contrary, it is to be interpreted as defining a position between these extremes which combines a fair protection for the patentee with a reasonable degree of certainty for third parties.

45. If a patentee were at liberty to narrow down the proper meaning of his claims by importing limitations which are not to be found in those claims on a purposive construction, but only in the body of the specification, there would not be 'a reasonable degree of certainty for third parties'. Many years ago, before the modern approach to claim drafting was devised, there was an inordinate amount of extremely expensive litigation in this country because each side sought to put a gloss on the vague language of the type of claim used in those days by referring to this or that passage in the body of the specification. Beyond question the expense and uncertainty must often have served to deter legitimate competitors from coming into the market in the first place, to the detriment of industry. Nowadays a reasonable degree of certainty for third parties is not merely a laudable aim, it is a mandatory requirement of the law.

46. Nor is this approach to construction necessary in order to achieve 'a fair protection for the patentee', also mandatory. If the patentee considered that the concept was a significant feature of his invention he could have chosen reasonably indicative language to express it in the first place. Moreover, if he has omitted to do so because of a mistake on his part in will normally be open to him to apply to amend his patent in order to cure the defect.

47. Therefore, and despite the skill with which Mr Tritton put his case, I hold that claim 1 would be satisfied by the presence of features (a)-(e) above. It does not require 'an integrated uninterrupted "software data" link between the design and the cutting and embroidery machines'. No doubt it would be more convenient to program a single computer, and to perform all the necessary analytical work thereon, and to use it to command both the embroidery machine and the cutting machine. But a manufacturer would not avoid infringement of claim 1 by using two independent computers and correctly deriving the cutting and embroidery command data separately from the same design drawing. If I am right in thinking so, claim 1 is by concession invalid for obviousness.

48. In coming to that conclusion I have not ignored an argument which, I believe, may best be put in the following way. It is that the skilled reader would not read the claim 1 in the manner I have summarised above, because it would be apparent to him that it (the absence of an integrated uninterrupted link) would have a materially adverse effect on the way the invention works; hence he would suppose that the patentee had intended Mr Tritton's feature to be an essential feature of the invention. This is analogous to the Catnic/Improver tests (conveniently summarised in the judgement of Hoffmann J in Improver Corporation v. Remington Consumer Products Ltd [1990] FSR 181, 188-9) which are habitually used by the courts in order to arrive at a purposive construction of patent claims, although those tests are normally applied in the context of a claim which is said to be wider than its primary, literal or acontextual meaning.

49. This argument is founded on two propositions: (1) that the skilled reader would have appreciated that a problem with the prior art processes was that it was difficult to cut out the appliqué pieces to match the outline of the 'fixing' stitch; and (2) that he would have appreciated (after reading the patent, anyway) that the cause of this problem was a failure to integrate the derivation of the cutting command data and the embroidery command data, respectively.

50. In my judgement the first proposition is by no means entirely true and the second is wrong. I believe pieces could be cut to the right shape and size with a die for use on a Multi-head machine. It may be that it was sometimes difficult to produce appliqué cut-out elements of intricate shape to the required degree of precision if one used a die.

51. But insofar as there was a problem about matching the data which defined the outline of the cut-out appliqué element to the data which prescribed the outline stitch, I do not believe the solution consisted in providing an 'uninterrupted "software data" link'. Why should this make the difference? It seems to me that if a computer can do this with the right software, as contemplated in the patent, then two computers (one linked to the cutting machine and the other linked to the embroidery machine), each computer being provided with the right software and access to identical copies of the design drawing, would do so with equal success. In sum, I am not at all satisfied that the skilled reader of the patent would read claim 1 in the manner desired by the Claimant.

52. It may be that the APS software devised by the Claimant possessed some attribute which enabled it to work with success where other, inferior software would have failed. Such indeed was the evidence of Mr Widincamp (Day 2, pages 187-193). It was his job to promote the Claimant's system and to deal with customers' technical queries. I shall quote some extracts from this testimony.

Q. ... [A] client gives you a design ... how does he do that? On paper?

A. The majority of the artwork was camera-ready...

Q. And that design will have with it something that defines the stitches he wants?

A. Yes sir.

Q. And something that defines the outline?

A. There is quite a bit of artistic interpretation ... One of the inherent problems with embroidery – working from camera-ready – was that you had to interpret camera-ready into what something would look lie in embroidery. It's an artistic interpretation. And therefore the appliqué and the stitching and the effects of the stitching was really subject to the artist's opinion of what would look best.

Q. And did this new software do that well?

A. Yes, sir.

Q. Is that the reason for its success?

A. Yes, sir.

.....

Q. Okay. So I've done the process you've just described; digitised my stitch data; got it in your machine: how do I get the cutting data?

A. What happened was that the cutting data – what made this revolutionary –

Q. We don't want to know what made it revolutionary. How did I get the cutting data?

A. The cut line was derived from the stitch data.

Q. ... Okay. So I've got stitch data on my machine. I've got it printed up. I've got it on my computer screen. How do I tell what of that I want to use as cutting data?

A. The cutting data would be extracted from the design and, you know, you would get the stitch data from the design, and the cutting data from the design, off the same screen.

Q. Was that something that, basically, was up to the software designer, not within your field, to know how to do that?

A. No, it was completely in the software.

Q. Right. If you yourself were asked how to do it, you wouldn't have known how to do it?

A. No, sir ... I'm not a software engineer.

Q. Okay, we'll leave it then. Pretend I'm a customer... I give you my picture. You return a series of cutout appliqué patches, and you return a disc with stitch command data on [it] for me?

A. Hmm.

Q. Why do I care how you have cut them out?

A. It is important to understand, again, that at this time there was nothing available with the precision that we could get with the file being from the same source and the laser cutter combination. There was nothing that could approach the precision.

Q. Nothing that could approach that precision. You have told me, "As part of the process, I have to feed all this information manually."

A. Yes.

Q. And when I make a die, I have to create it manually?

A. Yes.

Q. So why couldn't my returned shapes have been die cut?

A. Simply because, in the process of making a die, there is some distortion that happens. You can't bend [i.e. with precision]. There are certain angles and certain complexities that can't be achieved in a die. When you bend these angles, these dies are bent by hand and, as they are bent, the angles change slightly, so that you stitches don't fit the die [cut piece] as accurately.

Q. Okay. So the advantage is that the shapes you return to me can be of a more intricate shape than die cutting shapes. And this is because they are laser cut?

A. It's because of the [computer] file. Because the cutting data is exactly a match to the stitching data. That's the important thing...

Q. This service that we're talking about, the one before 1992, was commonplace?

A. Yes, sir. It's important to understand that appliqué is a very small part of embroidery.

Q. Well known in the trade?

A. Yeah, I would imagine so, yes.

Q. Okay. So the only difference between your service and all the prior art services – all the services that were done before – was, as you would say, how you derived the design and the fact that you laser cut it?

A. I would say that it how we derived the design and the fact that the precision was there that wasn't there before.

53. In my judgement the success of the Claimant's commercial process was caused by the use of a precision laser cutter which could be programmed with the necessary command data. That success may have been enhanced by its provision of a purpose-built laser cutter which it sold to customers. Also, there may have been something special about the Claimant's software – something not disclosed in the patent – which ensured that the design drawing would be well interpreted. But what I do not accept is that the cause of the proper matching of the cutting and hold-down stitch command data was the presence of an 'uninterrupted "software data" link' or 'the file being from the same source'. I reject this as technically implausible.

54. Of course, if the cutting data and the hold-down stitching data were ill-matched this could have caused problems; but where does this take us? Employing the same general logic as is urged by the Claimant, I would then have to interpret the expression 'analysing the design' in claim 1 as analysing the design correctly to avoid such ill-matching. This may be so, but does not give us Mr Tritton's extra feature.

55. In case I am wrong, however, I shall go on to consider the validity of claim 1 on Mr Tritton's construction. Before I do I shall briefly indicate why I believe he was quite right to concede that the invention according to the wider interpretation of claim 1 was obvious.

Obviousness of the invention according to the wider interpretation of claim 1.

56. Before May 1992 appliqué pieces were either trimmed by hand after being stitched to the base fabric; or they were produced by means of a die which had to be made to order, this being a separate trade. Thus the process was either labour-intensive or it involved delay. In my judgement, it would be a natural and routine part of modern management technique to identify where time was being wasted in the existing (die) process, and to enquire: "Is there anything we can do to avoid this? Is more modern machinery available which could help us improve?" A technical manager given the task of reporting on those questions would have found out by making normal enquiries that there existed computer-controlled, precision laser cutters which would have answered the purpose – indeed I believe he would know it already. After all, it had been described as "the wave of the future" in the last sentence of a practical textbook published shortly before the priority date: the skilled team may or may not have read the book or that sentence, but in my judgement it correctly represented the sentiment at the time. The command data for the laser cutter would inevitably have had to be derived from the design drawing. In my judgement nothing in the foregoing would have involved any inventive step.

57. It might well be that, having arrived at the concept, a manufacturer would have decided that laser cutters were at that time too expensive to justify the investment. He might have said "We had better wait until the price comes down". Or, for all that I know, he might have been a manufacturer of laser cutters himself, and have said: "Here is a business opportunity. Let us spend money developing a small, cheap laser cutter tailor-made for the appliqué people". But those are questions of mere economics or business judgement. It may have been astute business judgement on the part of the Claimant – and I suspect that it was – to develop a laser cutter which was particularly apt for the purpose, in terms of cost or otherwise, thus filling a gap in the market. Instructions for making such a laser cutter are not contained in the patent. It has nothing to do with whether the concept defined in claim 1 was obvious. In my judgement it clearly was. What has to be determined is whether what is now claimed as inventive would have been obvious at the priority date, not whether it would have appeared commercially worthwhile to exploit it at the time: Windsurfing case at 72; Hallen Co v. Brabantia (UK) Ltd [1991] RPC 195, 213, C.A.; Thermos Ltd v. Isola Ltd (1910) 27 RPC 388, 398-9.

Obviousness of the invention according to the narrower interpretation.

58. Assuming, however, that Mr Tritton's construction of claim 1 is correct, I should decide whether the invention thus defined was or was not obvious. I regard it as my duty to decide this even though I do not agree with that construction. This is because the point has been raised and argued and it could be relevant to amendment of claim 1 of the patent, see the Windsurfing case at pages 81-82.

59. Obviousness has been described as a 'jury' question and a useful approach which is commonly adopted in answering it was formulated by Oliver LJ in Windsurfing International BV v. Tabur Marine (G.B.) Ltd [1985] RPC 59, 73-4. I have already adopted that approach in coming to my conclusion above although I have not set out the formulation expressly, since the invention according to the wider interpretation of claim 1 is obvious by concession. However I shall do so now. The passage reads as follows:

There are ... four steps which require to be taken in answering the jury question. The first is to identify the inventive concept embodied in the patent in suit. Thereafter, the court has to assume the mantle of the normally skilled but unimaginative addressee in the art at the priority date and to impute to him what was, at that date, common general knowledge in the art in question. The third step is to identify what, if any, differences exist between the matter cited as being "known or used" and the alleged invention. Finally, the court has to ask itself whether, viewed without any knowledge of the alleged invention, those differences constitute steps which would have been obvious to the skilled man or whether they require any degree of invention.

60. Under the Patents Act 1977 the test now is, not matter which was 'known or used, but matter which has been at the priority date "made available to the public". Here the Defendants rely on what was common general knowledge.

a) The inventive concept.

I shall here assume, following Mr Tritton's opening submissions as revised in his closing speech, that the inventive concept was to integrate in a computer-controlled process the cutting of appliqué material with the embroidery and stitching of the appliqué material to the base fabric so that the effective co-ordinate data of the outline of the appliqué can be used both to cut out the appliqué by a computer-controlled machine and to stitch it to the base fabric around the outline by a computer-controlled embroidery machine.

b) The skilled addressee and common general knowledge.

It is well settled that the notional skilled addressee of a patent may be, where appropriate, a team of persons, each possessed of a specialist skill. Mr Tritton submitted that the skilled addressee would be one involved in the field of industrial embroidery, with a knowledge of such machines as were used in that industry. He would know that there existed computer-controlled fabric-cutting machines. He would have on his team one skilled in computer programming. In my judgement Mr Tritton is right. They would also know that precision laser cutters, computer-controlled, were available. The patent itself presupposes that they were, and in my judgement the skilled team would have found out by making routine enquiries or, more probably, have known it already.

The skilled team would know about machines of the Schiffli and Multi-head types. In any case in May 1992 they would know that embroidery machines were nowadays computer-controlled – the patent itself says so. They would know that in order to derive the requisite stitch command data a representation of the design would be entered into a computer memory, either by means of a design program (CAD) or by some kind of scanning process such as video – again the patent says so. They would know that, in order to instruct the computer-controlled cutting machine, appropriate command data would have to be generated by some means or another, this being a task for the software specialist.

61. They would know that appliqué elements were made either by producing oversize elements and trimming them by hand after applying the stitches, or by cutting them out to size from cloth by means of a purpose-made die.

62. Although the computer protocols which were used to command the then-available embroidery and cutting machines, respectively, might not be the same, any competent computer specialist would know that one computer format may be translated into another, preferably and usually by automated data processing means. Thus if the team arrived at the concept of the invention, any such incompatibilities would not deter it at all, in my judgement; nor would they serve as an effective deterrent to arriving at the concept in the first place.

63. I hold that all of the above matters were common general knowledge in the art. They would have or ought to have been known to any competent person skilled in the art. Or, at the very least, they could and would have been ascertained by any competent technical management team who asked themselves "How can we improve the existing process?" and pursued normal, routine enquiries.

64. Although the members of the team would have been skilled persons, I am to assume that they were not imaginative, and I shall do so.

c) Differences between what was known and the alleged invention.

In order to arrive at the alleged inventive concept the skilled team would have to think of pre-cutting the appliqué pieces to the right shape and size by the use of a computer-controlled cutting machine – in practice, a laser cutter – instead of by the use of a die. They would also have to think of the idea of creating a single computer file containing data which would serve to command both the embroidery machine and the cutting machine according to the 'integrated' concept espoused by Mr Triton.

d) Whether differences were obvious steps.

The Court of Appeal in the Windsurfing case said that this evaluation must be made 'viewed without any knowledge of the alleged invention'. This is very important, and the courts have repeatedly warned against the danger of being wise after the event. For example, in Vickers Sons & Co Ltd v. Siddell (1890) 7 RPC 292, 304-5, Lord Herschell said:

But experience has shown that not a few inventions, some of which have revolutionised the industries of this country, have been of so simple a character that when once they were made known it was difficult to understand how the idea had been so long in presenting itself, or not to believe that they must have been obvious to anyone.

65. I have already explained why I believe it would have been obvious to consider the use of the latest computer-controlled laser cutters. Their evident advantages were precision and the ability to produce the required cutting outline by programming, in contrast to the cumbersome hand or die methods. Indeed, unless technical management were to have deliberately shut their eyes to the possibility of adopting modern improvements such consideration was not only obvious, but practically inevitable.

66. In my judgement the simplest manner of providing the laser cutter with the requisite command data was also obvious. In order to acquire cutting command data it was necessary to digitise the appropriate parts of the design drawing. It was already standard practice to digitise ('punch') the drawing either by creating it directly on the computer itself, or by scanning it into the memory by well known means. For forensic reasons I have had to consider a scenario in which the drawing was computer-analysed twice over, on two distinct computers, once to obtain the stitch command data and once to obtain the cutting command data. But this is faintly absurd. Why would anyone go to that extra trouble and expense? The obvious thing to have done was to provide a single computer and a single software package that would perform both steps. The hold-down stitches would have to be applied to the margin of the cut-out appliqué piece, typically in a zigzag pattern. (If this is not true Mr Tritton's point fails anyway.) Thus to the extent that the stitching line and of the cutting line were the same, the same line would have to serve as the starting place for both. Realistically, what else was the skilled addressee to do? Making every allowance, it seems to me that there was no inventive step.

67. Dr White gave supplementary evidence in chief which was not in his report. He told me that in 1993 or 1994 Tajima, a prominent manufacturer, introduced an embroidery machine for appliqué work which incorporated a laser cutter but which worked on a different principle. In essence, the machine automated the old hand-trimming process. An oversize piece would be stitched to the base fabric; then the machine frame would move to one side; then the laser beam would be focused so as to trim the appliqué piece to the correct size and shape (yet not cut so deeply as to cut through the base fabric); then the frame would move back again and the remainder of the embroidery stitches would be accomplished. The point of this evidence was to lay a foundation for an argument that the invention of claim 1 was not obvious since here was a reputable manufacturer doing it another way.

68. The fact that it could have been done in another way does not detract from the obviousness of the method of the patent. There were two starting places, the hand-trimming method and the method (better suited to large orders) according to which the piece was die-cut to the required shape in advance. It may be that – at the level of generality which I have described – the process used by the Tajima machine was obvious, but that it not a matter before me. I note in passing that, in any case, the Tajima machine derived the embroidery command data and the cutting command data in one integrated process, which is what I would have expected any sensible design team to have devised.

69. Even if it should appear at first sight that an alleged invention was obvious, the surrounding circumstances may show otherwise. Thus if there was a long-felt want for the invention and its introduction resulted in commercial success that would not have occurred absent the inventive concept, that is usually a fairly good indication that it was not obvious after all.

70. The Claimant's system has indeed been a commercial success, and it did satisfy a long-felt want in that its employment did away with the need to finish the appliqué pieces by hand-trimming, which was labour-intensive and time-consuming; or the need to have dies made to order, which not only must have imposed inconvenient delays but might have resulted in poor matching if the shapes were complex.

71. However, while I bear those facts in mind, the success of the Claimant's system is not attributable to the alleged inventive concept. In my judgement it was caused by (1) the advent of precision laser-cutters which were computer-controlled (2) possibly, the development by the Claimant of a laser-cutter (not described in the patent) which was particularly suited to the market (3) possibly, the excellence of the APS software created by the Claimant. If it were asked, why was not this done before, I would reply: because such computer-controlled laser cutters as were available before the priority date of the patent were very expensive. Nevertheless, at that date they were known to be, in Mr Schneider's words, the wave of the future. I do not believe the so-called 'secondary factors' present in this case refute the proper finding, which is that the invention of claim 1, on whatever construction, was obvious.

Conclusion.

72. Claim 1 is the only claim of the patent which is alleged to have been infringed by the Defendants yet to possess independent inventive significance. I have held that its subject-matter was obvious.

73. The Claimant did not suggest that if claim 1 were to be held invalid even on the narrow construction the defect could be cured by a suitable amendment of the patent which would result in the claim covering what the Defendants are doing; in any case, the Claimant did not suggest what that amendment might be or seek to persuade me that the (hypothetical) amendment would be effective for its purpose. See the Windsurfing case at pages 81-82. In those circumstances I believe my proper course is to dismiss the claim.

74. There is a counterclaim for, amongst other things, an order that the patent be revoked. Clearly it would not be appropriate to revoke the patent as a whole seeing that the Defendants have not sought to persuade me that the other claims of the patent are invalid. I shall consider what other relief is appropriate after considering the parties' submissions. In asking for such relief the parties should bear in mind that proceedings (particularly in this court) are meant to be cost-effective. If the Claimant wishes to amend its patent it should give consideration to what would be the appropriate forum for that purpose, seeing that according to my judgement the Defendants would no longer be involved. Unless a special application is made for oral proceedings, supported on reasonably persuasive grounds, I propose to consider the parties' submissions on the basis of written arguments alone.